Top Banner
’%&" ,GGN<E 7@IHKM HA 2Q?KHB@HEHBC> .HG?CMCHGL" ;<M@K 7@JNCK@F@GML" 8NIIEC@L <G? /CLIHLCMCHG 8<GM< 5<KC< 9<EE@Q 5<G<B@F@GM ,K@< ,IKCE" ’%&# Luhdorff and Scalmanini Consulting Engineers
143

GGN .HG?CMCHGL ;

Feb 03, 2021

Download

Documents

dariahiddleston
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
  • '%&" ,GGN .HG?CMCHGL"

    ;

  • i

    Table of Contents

    Page

    1. Introduction........................................................................................................................11.1 Physical Setting.................................................................................................11.2 Previous Studies................................................................................................21.3 SMVMA Monitoring Program .........................................................................21.4 Additional Monitoring and Reporting Programs ..............................................41.5 Report Organization..........................................................................................4

    2. Hydrogeologic Conditions.................................................................................................52.1 Groundwater Conditions...................................................................................5

    2.1.1 Geology and Aquifer System..............................................................52.1.2 Groundwater Levels............................................................................82.1.3 Groundwater Quality ........................................................................11

    2.2 Twitchell Reservoir Operations ......................................................................132.2.1 Reservoir Stage and Storage .............................................................132.2.2 Reservoir Releases ............................................................................14

    2.3 Streams............................................................................................................152.3.1 Discharge ..........................................................................................152.3.2 Surface Water Quality.......................................................................17

    2.4 Climate............................................................................................................192.4.1 Precipitation ......................................................................................192.4.2 Evapotranspiration ............................................................................20

    3. Water Requirements and Water Supplies.....................................................................213.1 Agricultural Water Requirements and Supplies .............................................21

    3.1.1 Land Use ...........................................................................................213.1.2 Applied Crop Water Requirements...................................................223.1.3 Total Agricultural Water Requirements............................................233.1.4 Agricultural Groundwater Pumping .................................................24

    3.2 Municipal Water Requirements and Supplies.................................................243.2.1 Municipal Groundwater Pumping.....................................................243.2.2 Imported Water .................................................................................253.2.3 Total Municipal Water Requirements...............................................27

    3.3 Total Water Requirements and Supplies..........................................................27

    4. Water Disposition ............................................................................................................294.1 Agricultural Return Flows ..............................................................................294.2 Treated Municipal Waste Water Discharge....................................................304.3 Agricultural Drainage .....................................................................................324.4 Exported Water (Intra-Basin Transfer)...........................................................33

  • ii

    Table of Contents, cont.

    Page

    5. Conclusions and Recommendations...............................................................................375.1 Conclusions.....................................................................................................37

    5.1.1 Hydrogeologic Conditions ................................................................375.1.2 Water Requirements, Supplies, and Disposition...............................375.1.3 Stipulation ........................................................................................39

    5.2 Recommendations...........................................................................................40

    6. References ........................................................................................................................43

    Appendices

    Appendix A SMVMA Monitoring ProgramAppendix B Historical Groundwater Quality, Coastal Monitoring WellsAppendix C 2015 Land Use Interpretation Data and Image InventoryAppendix D Estimated Historical Return Flows, Waste Water Treatment PlantsAppendix E Calculation of Landscape Irrigation Return Flows, Annually from 2008

  • iii

    List of Figures

    1.1-1 Santa Maria Valley Groundwater Basin and Management Area

    2.1-1a Generalized Geologic Map with Cross Section Locations2.1-1b Longitudinal Geologic Cross Section, A-A’2.1-1c Transverse Geologic Cross Section, B-B’2.1-2 Historical Groundwater Levels2.1-3a Contours of Equal Groundwater Elevation, Shallow Zone, Early Spring 20152.1-3b Contours of Equal Groundwater Elevation, Shallow Zone, Late Spring 20152.1-3c Contours of Equal Groundwater Elevation, Shallow Zone, Fall 20152.1-3d Contours of Equal Groundwater Elevation, Deep Zone, Early Spring 20152.1-3e Contours of Equal Groundwater Elevation, Deep Zone, Late Spring 20152.1-3f Contours of Equal Groundwater Elevation, Deep Zone, Fall 20152.1-4a Seasonal Groundwater Level Fluctuations, Northern Santa Maria Valley2.1-4b Seasonal Groundwater Level Fluctuations, Southern Nipomo Mesa2.1-5 Historical Groundwater Quality

    2.2-1a Historical Stage and Storage, Twitchell Reservoir2.2-1b Historical Releases, Twitchell Reservoir

    2.3-1a Historical Surface Water Discharge, Cuyama River and Twitchell Reservoir2.3-1b Historical Stream Discharge, Sisquoc River2.3-1c Historical Stream Discharge, Santa Maria River2.3-1d Historical Stream Discharge, Orcutt Creek2.3-2a Historical Surface Water Quality, Sisquoc River2.3-2b Historical Surface Water Quality, Orcutt Creek

    2.4-1 Historical Precipitation, Santa Maria Airport2.4-2 Historical Reference Evapotranspiration, CIMIS Stations

    3.1-1a Agricultural Land Use, 20153.1-1b Historical Distribution of Irrigated Acreage, by Crop Category3.1-1c Historical Agricultural Acreage and Groundwater Pumping

    3.2-1a Historical Municipal Groundwater Pumping3.2-1b Historical State Water Project Deliveries3.2-1c Historical Municipal Water Requirements

    3.3-1 Historical Total Water Requirements

  • iv

    List of Tables

    2.3.1 Selected General Mineral Constituent Concentrations, Santa Maria Valley Streams

    2.4-1 Precipitation Data, 2015, Santa Maria Airport2.4-2 Reference Evapotranspiration and Precipitation Data, 2015, SMVMA CIMIS Stations

    3.1-1a Distribution of Irrigated Acreage, 20153.1-1b Historical Distribution of Irrigated Acreage3.1-1c Applied Crop Water Requirements and Total Agricultural Water Requirements, 2015

    3.2-1a Municipal Groundwater Pumpage, 20153.2-1b Municipal State Water Project Water Deliveries, 20153.2-1c Historical Municipal Water Requirements and Supplies

    3.3-1a Total Water Requirements and Supplies, 20153.3-1b Recent Historical Total Water Supplies

    4.1-1 Applied Crop Water Requirements, Total Agricultural Water Requirements and ReturnFlows, 2015

    4.2-1 Treated Municipal Waste Water Discharge, 20154.2-2 Estimated Recent Historical Return Flows from WWTPs and Landscape Irrigation

    4.3-1 Water Requirements, Supplies, and Amounts Delivered, Current and ProjectedConditions, Wholesale Water Supply Agreement, May 7, 2013

    5.1-1 Summary of 2015 Total Water Requirements, Water Supplies, and Disposition

  • v

    Acronyms and Abbreviations

    af, afy, af/ac acre-feet, acre-feet per year, acre-feet/acre

    AW applied water

    CCAMP Central Coast Ambient Monitoring Program

    CCRWQCB Central Coast Regional Water Quality Control Board

    CCWA Central Coast Water Authority

    CIMIS California Irrigation Management Information System

    DU Distribution Uniformity

    DWR California Department of Water Resources

    ET evapotranspiration

    ETaw, ETc, ETo ET of applied water, ET of the crop, Reference ET

    GIS Geographic Information System

    GPD Gallons per day

    GSWC Golden State Water Company

    Kc crop coefficient

    Laguna CSD Laguna County Sanitation District

    LSCE Luhdorff & Scalmanini, Consulting Engineers

    mg/l milligrams per liter

    MOU Memorandum of Understanding

    Nipomo CSD Nipomo Community Services District

    NMMA (TG) Nipomo Mesa Management Area (Technical Group)

    NO3-NO3 nitrate-as-nitrate

    NOAA National Oceanic and Atmospheric Administration

    PE effective precipitation

    SBCWA Santa Barbara County Water Agency

    SCWC Southern California Water Company

    SLODPW San Luis Obispo County Department of Public Works

    SMVMA Santa Maria Valley Management Area

    SMVWCD Santa Maria Valley Water Conservation District

    SWP State Water Project

    SWRCB State Water Resources Control Board

    TMA Twitchell Management Authority

    UCCE University of California Cooperative Extension

    USGS United States Geological Survey

    umho/cm micromhos per centimeter

    WIP Waterline Intertie Project

    WWTP waste water treatment plant

  • 1

    1. Introduction

    This annual report of conditions in the Santa Maria Valley Management Area, for calendar year2015, has been prepared to meet the reporting conditions of the June 30, 2005, Stipulationentered by the Superior Court of the State of California, County of Santa Clara, in the SantaMaria Valley Groundwater Basin litigation. The Stipulation divided the overall Santa MariaValley Groundwater Basin into three management areas, the largest of which overlies the mainSanta Maria Valley (the Santa Maria Valley Management Area, or SMVMA) and is the subjectof this report. The other two management areas, the Nipomo Mesa Management Area (NMMA)and the Northern Cities Management Area, are addressed in separate annual reports prepared byothers.

    The Stipulation, approved and implemented in 2008, specifies that monitoring shall be sufficientto determine groundwater conditions, land and water uses, sources of water supply, and thedisposition of all water supplies in the Basin. Annual Reports for the SMVMA are to summarizethe results of the monitoring and include an analysis of the relationship between projected waterdemand and supply. The Stipulation was preserved in the California Court of Appeal (SixthAppellate District) Decision of November 21, 2012, and in the Superior Court of the State ofCalifornia (County of Santa Clara) Final Judgment of April 23, 2014. Thus, the PhysicalSolution criteria for monitoring and managing groundwater in the basin remain.

    In accordance with the Stipulation, this report on the SMVMA provides a description of thephysical setting and briefly describes previous studies conducted in the groundwater basin,including the long-term monitoring program developed for the SMVMA. As reported herein, theTwitchell Management Authority (TMA) commissioned the preparation of a monitoring programfor the SMVMA in 2008, and its complete implementation is expected to provide the data withwhich to fully assess future conditions. This report describes hydrogeologic conditions in themanagement area historically and through 2015, including groundwater conditions, TwitchellReservoir operations, and hydrologic and climatic conditions. As with all previous annualreports (2008 through 2014), the water requirements and supplies for agricultural and municipaluses are accounted, as are the components of water disposition in the SMVMA. Conclusionsdrawn regarding water resource conditions are discussed, including any finding of severe watershortage, which is concluded to not be the case through 2015. Finally, recommendations areprovided with regard to the intra-basin transfer of water from the SMVMA to the NMMA,possible enhancement of groundwater recharge, expanded assessment of water resourceconditions, and continued implementation of the monitoring program for the SMVMA.

    1.1 Physical Setting

    The Santa Maria Valley Management Area (SMVMA) includes approximately 175 square milesof the Santa Maria Valley Groundwater Basin in northern Santa Barbara and southern San LuisObispo Counties, as shown by the location map of the area (Figure 1.1-1). The SMVMAencompasses the contiguous area of the Santa Maria Valley, Sisquoc plain, and Orcutt upland,and is primarily comprised of agricultural land and areas of native vegetation, as well as theurban areas of Santa Maria, Guadalupe, Orcutt, Sisquoc, and several small developments.

  • 2

    Surrounding the SMVMA are the Casmalia and Solomon Hills to the south, the San RafaelMountains to the southeast, the Sierra Madre Mountains to the east and northeast, the NipomoMesa to the north, and the Pacific Ocean to the west. The main stream is the Santa Maria River,which generally flanks the northern part of the Santa Maria Valley; other streams includeportions of the Cuyama River, Sisquoc River and tributaries, and Orcutt Creek.

    1.2 Previous Studies

    The first overall study of hydrogeologic conditions in the Santa Maria Valley described thegeneral geology, as well as groundwater levels and quality, agricultural water requirements, andgroundwater and surface water supplies as of 1930 (Lippincott, J.B., 1931). A subsequentcomprehensive study of the geology and hydrology of the Valley also provided estimates ofannual groundwater pumpage and return flows for 1929 through 1944 (USGS, Worts, G.F.,1951). A followup study provided estimates of the change in groundwater storage duringperiods prior to 1959 (USGS, Miller, G.A., and Evenson, R.E., 1966).

    Several additional studies have been conducted to describe the hydrogeology and groundwaterquality of the Valley (USGS, Hughes, J.L., 1977; California CCRWQCB, 1995) and coastalportion of the basin (California DWR, 1970), as well as overall water resources of the Valley(Toups Corp., 1976; SBCWA, 1994 and 1996). Of note are numerous land use surveys(California DWR, 1959, 1968, 1977, 1985, and 1995) and investigations of crop water use(California DWR, 1933, and 1975: Univ. of California Cooperative Extension, 1994; Hanson, B.,and Bendixen, W., 2004) that have been used in the estimation of agricultural water requirementsin the Valley. Recent investigation of the Santa Maria groundwater basin provided anassessment of hydrogeologic conditions, water requirements, and water supplies through 1997and an evaluation of basin yield (LSCE, 2000).

    1.3 SMVMA Monitoring Program

    In accordance with the Stipulation, a monitoring program was initially prepared in 2008 toprovide the fundamental data for ongoing annual assessments of groundwater conditions, waterrequirements, water supplies, and water disposition in the SMVMA (LSCE, 2008). As a basisfor designing the monitoring program, historical data on the geology and water resources of theSMVMA were compiled to define aquifer depth zones, specifically a shallow unconfined zoneand a deep semi-confined to confined zone, into which a majority of monitored wells wereclassified based on well depth and completion information. Assessment of the spatialdistribution of the wells throughout the SMVMA, as well as their vertical distribution within theaquifer system, provided the basis for designation of two well networks, one each for the shallowand deep aquifer zones. All network wells are to be monitored for groundwater levels, with asubset of those wells to be monitored for groundwater quality. Those wells with inconclusivedepth and completion information were originally designated as unclassified wells; in 2009 and2013, review of groundwater level and quality records allowed classification of some wells intothe shallow or deep aquifer zones. Accordingly, the monitoring program was revised in 2009and 2013 to reflect those minor changes to the well networks.

  • 3

    Also to be monitored are surface water conditions, specifically Twitchell Reservoir releases,stage, and storage, and stream discharge and quality, and climatic conditions, specificallyprecipitation and reference evapotranspiration data.

    In addition to the hydrologic data described above, the monitoring program specifies those datato be compiled to describe agricultural and municipal water requirements and water supplies.These include land use surveys, to serve as a basis for the estimation of agricultural irrigationrequirements, and municipal groundwater pumping and imported water records, including anytransfers between purveyors.

    Lastly, the monitoring program for the SMVMA specifies water disposition data be compiled,including treated water discharged at waste water treatment plants (WWTPs) and any waterexported or transferred from the SMVMA. As part of this accounting, estimation is to be madeof agricultural drainage from the SMVMA and return flows to the aquifer system. For reference,the SMVMA monitoring program is included in Appendix A.

    In order to complete this annual assessment of groundwater conditions, water requirements,water supplies, and water disposition in the SMVMA, the following data for 2015 were acquiredfrom the identified sources:

    - groundwater level and quality data: the US Geological Survey (USGS), the Santa MariaValley Water Conservation District (SMVWCD), the Technical Group for the adjacentNMMA (NMMA TG), the City of Santa Maria, and Golden State Water Company;

    - Twitchell Reservoir stage, storage, and release data: the SMVWCD and Santa BarbaraCounty Public Works Department;

    - surface water discharge and quality data: the USGS and the Central Coast RegionalWater Quality Control Board (CCRWQCB);

    - precipitation data: the National Weather Service of the National Oceanic andAtmospheric Administration (NOAA), California Department of Water Resources(DWR), and SMVWCD;

    - reference evapotranspiration and evaporation data: the California DWR, includingCalifornia Irrigation Management Information System (CIMIS), and SMVWCD,respectively;

    - agricultural land use data, aerial photography, and satellite imagery: Santa Barbara andSan Luis Obispo County Agricultural Commissioner’s Offices; US Department ofAgriculture (USDA); and USGS;

    - municipal groundwater pumping and imported water data: the City of Santa Maria, theCity of Guadalupe, and the Golden State Water Company; and

  • 4

    - treated municipal waste water data: the City of Santa Maria, the City of Guadalupe, theLaguna Sanitation District, and the CCRWQCB.

    1.4 Additional Monitoring and Reporting Programs

    In 2014, the TMA was designated by the Calif. DWR as the Monitoring Entity for the SMVMAunder DWR’s California Statewide Groundwater Elevation Monitoring (CASGEM) Program.Compliance with the CASGEM Program requirements, which include at least semi-annualmonitoring and reporting of groundwater levels in a subset of shallow and deep wells alreadywithin the SMVMA Monitoring Program, is fulfilled by the TMA.

    Additionally, in 2016, groundwater resource planning and data reporting requirements under theCalif. DWR Sustainable Groundwater Management Program (SGMA) commenced. Since theSMVMA is part of an adjudicated basin, the Calif. DWR considers it already managed by theCourt and, thus, SGMA groundwater resource planning requirements do not apply. Theremaining SGMA requirements for reporting water resources data such as groundwater levels,groundwater pumping, and imported water amounts, were fulfilled by LSCE in its capacity asManagement Area Engineer under the Stipulation.

    1.5 Report Organization

    To comply with items to be reported as delineated in the Stipulation, this annual report isorganized into five chapters:

    - this Introduction;

    - discussion of Hydrogeologic Conditions, including groundwater, Twitchell Reservoir,surface streams, and climate;

    - description and quantification of Water Requirements and Water Supplies for the twooverall categories of agricultural and municipal land and water use in the SMVMA;

    - description and quantification of Water Disposition in the SMVMA; and

    - summary Conclusions and Recommendations related to findings regarding waterresource conditions in the SMVMA, in 2015 as well as historically, and recommendedactions related to water transfer, groundwater recharge, water resource assessment, andwater resource monitoring.

  • 09

    N/3

    4W

    10

    N/3

    4W

    09

    N/3

    5W

    09

    N/3

    3W

    10

    N/3

    3W

    10

    N/3

    5W

    09

    N/3

    2W

    10

    N/3

    2W

    11

    N/3

    2W

    11

    N/3

    3W

    11

    N/3

    4W

    11

    N/3

    5W

    09

    N/3

    1W

    10

    N/3

    1W

    11

    N/3

    1W

    10

    N/3

    6W

    08

    N/3

    5W

    08

    N/3

    4W

    08

    N/3

    3W

    08

    N/3

    2W

    08

    N/3

    1W

    11

    N/3

    6W

    09

    N/3

    6W

    08

    N/3

    6W

    0CB

    NK@

    &$&

    #&

    8<GM<

    5<KC<

    9<EE@Q

    1KHNG?P

    <M@

    K-

    <LC

    G<G?

    5<G<B@F

    @GM

    ,K@<L

    8<GM<

    5<KC<

    9<EE@Q

    5<G<B@F

    @GM

    ,K@<

    03

    61.5

    Mile

    s

    Le

    gen

    d

    Ma

    na

    ge

    men

    tA

    rea

    Bo

    un

    dary

    Gro

    un

    dw

    ate

    rB

    asin

    Bo

    un

    da

    ry

    SA

    NT

    AM

    AR

    IA

    VA

    LL

    EY

    SA

    NT

    AM

    AR

    IA

    VA

    LL

    EY

    NIP

    OM

    ON

    IP

    OM

    OM

    ES

    AM

    ES

    A

    NO

    RT

    HE

    RN

    NO

    RT

    HE

    RN

    CIT

    IE

    SC

    IT

    IE

    S

    SA

    NL

    UIS

    OB

    ISP

    OC

    OU

    NT

    Y

    SA

    NTA

    BA

    RB

    AR

    AC

    OU

    NT

    Y

  • 5

    2. Hydrogeologic Conditions

    Current and historical hydrogeologic conditions in the SMVMA, including groundwaterconditions, Twitchell Reservoir operations, and stream and climate conditions, are described inthe following sections of this Chapter.

    2.1 Groundwater Conditions

    To provide a framework for discussion of groundwater conditions, the geology of the SMVMA,including geologic structure and the nature and extent of geologic formations comprising theaquifer system, is described in the following section. Current groundwater levels are thendescribed in relation to historical trends in groundwater levels and flow directions in theSMVMA, as well as in context of Stipulation protocol for defining conditions of severe watershortage. Current and historical groundwater quality conditions are also discussed, includinggeneral groundwater quality characteristics as well as groundwater quality degradation,specifically due to elevated nitrate concentrations.

    2.1.1 Geology and Aquifer System

    The SMVMA is underlain by unconsolidated alluvial deposits that comprise the aquifer system,primarily gravel, sand, silt and clay that cumulatively range in thickness from about 200 to 2,800feet. The alluvial deposits fill a natural trough, which is composed of older folded andconsolidated sedimentary and metamorphic rocks with their deepest portions beneath the Orcuttarea. The consolidated rocks also flank the Valley and comprise the surrounding hills andmountains; typically, the consolidated rocks do not yield significant amounts of groundwater towells. The geologic formations comprising the alluvial deposits and the geologic structurewithin the study area are illustrated in a generalized geologic map (Figure 2.1-1a) and twogeologic cross sections (Figures 2.1-1b and 2.1-1c).

    The alluvial deposits are composed of the Careaga Sand and Paso Robles Formation (Fm.) atdepth, and the Orcutt Fm., Quaternary Alluvium, and river channel, dune sand, and terracedeposits at the surface (USGS, Worts, G.F., 1951). The Careaga Sand, which ranges in thicknessfrom about 650 feet to a feather edge, is identified as being the lowermost fresh water-bearingformation in the basin (DWR, 1970), resting on the above-mentioned consolidated rocks(specifically, the Tertiary-aged Foxen Mudstone, Sisquoc Fm., and Monterey Shale and theJurassic/Cretaceous-aged Franciscan Fm., descriptions of which may be found in USGS, Worts,G.F., 1951). Overlying the Careaga Sand is the Paso Robles Fm., which comprises the greatestthickness of the alluvial deposits (from about 2,000 feet to a feather edge); the thickest portion ofthis formation is located beneath the Orcutt area. Both the Careaga Sand and Paso Robles Fm.underlie the great majority of the SMVMA (see Figures 2.1-1b and 2.1-1c). The Careaga Sand ismainly composed of white to yellowish-brown, loosely-consolidated, massive, fossiliferous,medium- to fine-grained sand with some silt and is reported to be predominantly of marine origin(USGS, Worts, G.F., 1951). The Paso Robles Fm. is highly variable in color and texture,generally composed of yellow, blue, brown, grey, or white lenticular beds of: boulders andcoarse to fine gravel and clay; medium to fine sand and clay; gravel and sand; silt; and clay

  • 6

    (USGS, Worts, G.F., 1951). This formation is reported to be primarily fluvial (stream-laid) inorigin and there is no areal correlation possible between the individual beds, with the exceptionof a coarse basal gravel of minor thickness in the Santa Maria Valley oil field, generally in thesoutheast part of the SMVMA.

    Above the Paso Robles Fm. and comprising the Orcutt Upland is the Orcutt Fm., which istypically about 160 to 200 feet thick; in the remainder of the SMVMA, the Paso Robles Fm. isoverlain by the Quaternary Alluvium, which comprises the majority of the Valley floor and istypically about 100 to 200 feet thick. Further north in the adjacent NMMA, the Paso Robles Fm.is overlain by the Older Dune Sand, which comprises the Nipomo Mesa and ranges in thicknessfrom approximately 400 feet to a feather edge. Along the northeast edge of the Sisquoc plain,the Paso Robles Fm. is overlain by terrace deposits approximately 60 feet thick. The Orcutt Fm.is composed of conformable upper and lower units (“members”), both reported to be mainly offluvial origin that become finer toward the coast. The upper member generally consists ofreddish-brown, loosely-compacted, massive, medium-grained clean sand with some lenses ofclay, and the lower member is primarily grey to white, loosely-compacted, coarse-grained graveland sand (USGS, Worts, G.F., 1951).

    The Quaternary Alluvium is also composed of upper and lower members that are reported to bemainly fluvial in origin. The composition of the upper member becomes progressively finertoward the coast, with boulders, gravel, and sand in the Sisquoc plain area; sand with gravel inthe eastern/central Valley area; sand with silt from the City of Santa Maria to a pointapproximately halfway to Guadalupe; and clay and silt with minor lenses of sand and gravelfrom that area westward. The lower member is primarily coarse-grained boulders, gravel andsand with minor lenses of clay near the coast. The Older Dune Sand is composed of loosely- toslightly-compacted, massive, coarse- to fine-grained, well-rounded, cross-bedded quartz sandthat is locally stained dark reddish-brown (California DWR, 1999). The terrace deposits, ingeneral, are similar in composition to the coarse-grained parts of the Quaternary Alluvium.

    The alluvial deposits comprising the aquifer system include thin, discontinuous clay lenses,without thick sections of clay, at depth; further, the deposits lack peat (organic) layers. Thus, thepotential is remote for deep land subsidence to occur as the deposits dewater during periods ofdeclining groundwater levels. There are no known reports of, nor the potential for, landsubsidence in the SMVMA, as noted in a recent technical report on the occurrence of subsidencethrough California (California DWR, 2014).

    Two geologic cross sections illustrate several points about the geologic structure and variableaquifer thickness throughout the SMVMA. Longitudinal geologic cross section A-A’ (seeFigure 2.1-1b) begins in the area near the mouth of the Santa Maria River, traverses the OrcuttUpland, and terminates in the Sisquoc plain area near Round Corral, immediately southeast ofthe SMVMA. It shows the relative thicknesses of the various geologic formations and theirgeneral “thinning” from the central valley area toward the Sisquoc plain. This cross section alsoshows the Quaternary Alluvium and Orcutt Fm., essentially adjacent to each other andcomprising the uppermost aquifer in the SMVMA, divided into the above-described upper andlower members.

  • 7

    Transverse geologic cross section B-B’ (see Figure 2.1-1c) begins in the Casmalia Hills,traverses the western portion of the Valley (near the City of Guadalupe) and the southernNipomo Mesa, and terminates at Black Lake Canyon. It shows the prominent asymmetricalsyncline (folding of the consolidated rocks and Paso Robles Fm.) within the SMVMA andadjacent NMMA, with the deepest portion of Paso Robles Fm. toward the southern edge of theSMVMA, gradually becoming thinner and more shallow toward the north where it extendsbeneath the NMMA. This cross section also shows that both the upper and lower members ofthe Quaternary Alluvium extend north to the Santa Maria River, but only the upper memberextends beyond the River to the southern edge of the Nipomo Mesa, and neither member extendsnorthward beneath the Mesa.

    Several faults have been reported to be located in the SMVMA and adjacent portion of theNMMA. The Santa Maria and Bradley Canyon faults, located in the Valley in the area betweenthe City of Santa Maria and Fugler Point (at the confluence of the Cuyama and Sisquoc Rivers toform the Santa Maria River), are concealed and they are reported to be northwest-trending, high-angle faults, that vertically offset the consolidated rocks, Careaga Sand, and Paso Robles Fm.,but not the overlying Quaternary Alluvium or Orcutt Fm. (USGS, Worts, G.F., 1951). TheOceano and Santa Maria River faults are of a similar nature (the latter fault also has a significantstrike-slip component of movement), but they are primarily located in the southern NipomoMesa. The maximum vertical offset on the Oceano fault is reported to be in the range of 300 to400 feet within the Careaga Sand and Paso Robles Fm.; on the other faults, the vertical offset isreported to be much less, within the range of 80 to 150 feet (USGS, Worts, G.F., 1951;California DWR, 1999). However, these faults do not appear to affect groundwater flow withinthe SMVMA, based on the review of historical groundwater level contour maps (USGS, Worts,G.F., 1951; LSCE, 2000).

    There is no known structural (e.g., faulting) or lithologic isolation of the alluvial deposits fromthe Pacific Ocean; i.e., the Quaternary Alluvium, Orcutt Fm., Careaga Sand, and Paso RoblesFm. aquifers continue beneath the Ocean. Thus, there is geologic continuity that permitsgroundwater discharge from the SMVMA to the Ocean, and the potential exists for salt water tointrude into the coastal (landward) portions of the aquifers if hydrologic conditions within themwere to change.

    The aquifer system in the SMVMA is comprised of the Paso Robles Fm., the Orcutt Fm., and theQuaternary Alluvium (USGS, Worts, G.F., 1951). The upper member of the QuaternaryAlluvium is consistently finer-grained than the lower member throughout the Valley. Further,the upper member becomes finer grained toward the Ocean such that it confines groundwater inthe lower member from the approximate area of the City of Santa Maria's waste water treatmentplant westward (approximately eight miles inland from the coast). The result of this has beensome artesian conditions in the western valley area (historically, flowing artesian wells werereported until the early 1940s in the westernmost portion of the Valley) (USGS, Worts, G.F.,1951). More recently, many wells belonging to local farmers in the western valley area,specifically in the Oso Flaco area, began flowing again in response to rising confinedgroundwater levels, such as during the winter of 1999.

  • 8

    Analysis of the geology, groundwater levels, and groundwater quality indicates that the aquifersystem varies across the area and with depth, and this variation was the basis for the shallow anddeep aquifer zone designations of the SMVMA monitoring program (LSCE, 2008). In thecentral and major portion of the SMVMA, there is a shallow unconfined zone comprised of theQuaternary Alluvium, Orcutt Fm., and uppermost Paso Robles Fm., and a deep semi-confined toconfined zone comprised of the remaining Paso Robles Fm. and Careaga Sand. In the easternportion of the SMVMA where these formations are much thinner and comprised of coarsermaterials, particularly in the Sisquoc Valley, the aquifer system is essentially uniform withoutdistinct aquifer depth zones. In the coastal area where the surficial deposits (upper members ofQuaternary Alluvium and Orcutt Fm.) are extremely fine-grained, the underlying formations(lower members of Quaternary Alluvium and Orcutt Fm., Paso Robles Fm., and Careaga Sand)comprise a deep confined aquifer zone.

    2.1.2 Groundwater Levels

    Groundwater levels within the SMVMA have fluctuated greatly since the 1920's, when historicalwater level measurements began, with marked seasonal and long-term trends, as shown by acollection of representative groundwater level hydrographs from various areas throughout theSMVMA (Figure 2.1-2). The areas are designated on Figure 2.1-2 for illustrative purposes only,and include the so-called Coastal, Oso Flaco, Central Agricultural, Municipal Wellfield,Twitchell Recharge, and Sisquoc Valley areas. The historical groundwater level hydrographsillustrate that widespread decline in groundwater levels, from historical high to historical lowlevels, occurred between 1945 and the late 1960's. The declines ranged from approximately 20to 40 feet near the coast, to 70 feet near Orcutt, to as much as 100 feet further inland (in the areajust east of downtown Santa Maria). Those declines were observed in both the shallow and deepaquifer zones, and are interpreted today to have been the combined result of progressivelyincreasing agricultural (and to a lesser degree, municipal) demand and long-term drier thannormal climatic conditions during that period.

    Since the late 1960’s, the basin has alternately experienced significant recharge (recovery) anddecline which, collectively, reflect a general long-term stability as groundwater levels in bothaquifer zones have fluctuated between historical-low and near historical-high levels overalternating five- to 15-year periods. Groundwater levels throughout the SMVMA have shownthis trend, but with different ranges of fluctuation (see Figure 2.1-2); and groundwater levelshave repeatedly recovered to near or above previous historical-high levels, including as recentlyas 2002. Shallow groundwater levels in the Sisquoc Valley fluctuated somewhat differently inthat they did not fully recover to historical high levels by 2002. In the primary areas of rechargealong the Santa Maria River, groundwater level fluctuations are greater in the shallow aquiferzone than the deep (see Twitchell Recharge Area and Central Agricultural Area hydrographs).Conversely, in the Municipal Wellfield and Coastal Areas, groundwater level fluctuations aregreater in the deep aquifer zone. Hydrographs from wells along the coastal portion of theSMVMA show that groundwater elevations have remained above sea level, with deep (confined)groundwater levels rising enough to result in flow at the ground surface, throughout the historicalperiod of record. The periodic groundwater level fluctuation since the late 1960's (with a long-term stability) have apparently been due to intermittent wet and dry climatic conditions, withnatural recharge during wet periods complemented by supplemental recharge along the Santa

  • 9

    Maria River from the Twitchell Reservoir project (since becoming fully operational in the late1960's). Long-term stability would also appear to be partially attributable to a general "leveling-off" of agricultural land and water use in the basin since the early to mid-1970’s, as furtherdescribed in Chapter 3.

    Most recently, groundwater levels in both the shallow and deep zones have been in a graduallydeclining trend since 2002, and have declined more rapidly since 2012. Particularly in light ofprevailing land use and water requirements, this overall groundwater level decline can beconsidered to be primarily due to the fact that Twitchell Reservoir releases, for in-streamsupplemental groundwater recharge, and Sisquoc River discharge, have been well below thehistorical average in most years since 2000. More specifically, there were no TwitchellReservoir releases in nine of the last 14 years, including in 2013, 2014, and 2015. Further, theSisquoc River discharge was well below average in almost all years since 2002, as will bediscussed in Section 2.2. The declining trend in shallow groundwater levels was slowed orreversed during years 2005-2006 and again in 2010-2011. During these short periods, releasesfrom Twitchell Reservoir, as well as discharge in the Sisquoc River, were above averagefollowing above-average rainfall periods. However, with continuing dry conditions, andparticularly the current severe drought that commenced in 2012, the declining trend ingroundwater levels across the SMVMA resumed.

    In 2015, groundwater levels in one deep well in the Twitchell Recharge Area, specifically well10N/33W-30G1, declined to just below the historical low water level observed in this well in1991 (see Figure 2.1-2). However, water levels in numerous shallow and deep wells in the areawith long historical records (covering previous drought periods) did not decline below theirrespective historical low. Thus, it appears the 2015 water levels in well 30G1 reflect a localizedlowering of water levels. In fact, shallow and deep groundwater levels across the great majorityof the SMVMA remain above historical low levels in 2015. This includes along the coast wheregroundwater levels are well above sea level, indicating that the conditions conducive to sea waterintrusion are absent. As such, the groundwater level conditions observed in 2015 in theSMVMA do not meet Stipulation provisions defining a condition of severe water shortage, aswill be discussed in Section 5.1.

    Groundwater beneath the SMVMA has historically flowed to the west-northwest from theSisquoc area toward the Ocean, and this remained the case during 2015 as illustrated by contourmaps of equal groundwater elevation for the shallow and deep aquifer zones (Figures 2.1-3athrough 2.1-3f). As in most years of study in the basin, a notable feature in the contour maps in2015 is the widening of groundwater level contours beneath the central-south and westernportions of the SMVMA that indicates a reduced (flatter) groundwater gradient in this area. Thislikely reflects the fact that the majority of aquifer system recharge derives from streamflow inthe Sisquoc and Santa Maria Rivers, specifically in the eastern portion of the SMVMA upstreamof Bonita School Crossing Road, and to a certain extent from streamflow in creeks draining theCasmalia and Solomon Hills (such as Orcutt Creek) along the southern portion of the SMVMA.This is supported by the presence of a reduced groundwater gradient in this area since at least1960 (USGS, Miller, G.A., and Evenson, R.E., 1966; USGS, Hughes, J.L., 1977; LSCE, 2000).

  • 10

    The reduced gradient likely also reflects ongoing groundwater pumping in and around themunicipal wellfield near the Santa Maria Airport and Town of Orcutt where numerous deepmunicipal water supply wells of the City of Santa Maria and the Golden State Water Company(GSWC), and nearby agricultural wells, operate. This is supported by the observance that, in thisarea, the groundwater gradient in the deep aquifer zone is more reduced (flatter) than in theshallow zone. Further, groundwater elevations in the deep zone are markedly lower than those inthe shallow zone in this area, with smaller differences in groundwater elevations between depthzones in other portions of the SMVMA. Importantly, while the reduced groundwater gradientnear the municipal well field has had the effect of slowing the movement of groundwater throughthat portion of the SMVMA, it has not stopped or reversed the direction of groundwater flow.

    Also notable from the contour maps is the overall seasonal difference in groundwater levelsacross the SMVMA between the spring and fall periods. A decline in water levels was observedbetween early March and late April (early and late spring contour maps, respectively), withadditional decline through early October. The timing and magnitude of the decline presumablyreflect that of the area-wide groundwater pumping associated with seasonal agriculturalirrigation, as well as the fact that stream discharge in the Sisquoc River and the associatedrecharge to the aquifer system were essentially nil in 2015. Importantly, during 2015, a seawardgradient for offshore groundwater flow was maintained, and coastal groundwater elevationsremained well above sea level (typically exceeding 15 feet, NAVD88), in both the shallow anddeep aquifer zones. Across the coastal boundary of the SMVMA, the offshore flow ofgroundwater in the shallow aquifer zone appears to have been reduced by the fall, with a portionflowing to the northwest beneath the southern coastal Nipomo Mesa (see Figure 2.1-3c).Similarly, the offshore flow of groundwater in the deep zone appears to have been reduced,although by early spring (see Figure 2.1-3d). Regarding this groundwater flow from theSMVMA toward the southern coastal Nipomo Mesa, it appears to either meet groundwaterpumping demands in the area or eventually flow offshore along the coastal boundary of theNMMA.

    Additional information about the seasonal fluctuation of groundwater levels in and near theSMVMA, in particular along its northern boundary with the NMMA near Oso Flaco Valley, isderived from hourly groundwater level measurements made since late 2013 by transducers intwo monitoring wells belonging to the San Luis Obispo County Department of Public Works(SLODPW) in that area. A groundwater level hydrograph for one well located in thenorthwestern edge of Santa Maria Valley (Figure 2.1-4a) illustrates how, in 2015, groundwaterlevels were highest but already gradually declining in early January, before reaching their lowestlevels in September and October. In that area, the overall seasonal decline was about 12 feet.Importantly, it can be seen that groundwater levels did not recover fully in 2015; the year-endlevels were seven feet lower than those observed at the beginning of the year.

    A groundwater level hydrograph for the second well, located in the southern central edge of theNipomo Mesa (Figure 2.1-4b) indicates very similar groundwater level fluctuations, withgroundwater levels highest (but already declining) in early January before reaching their lowestlevels in early October. An overall seasonal water level decline of about 20 feet was observedbetween January and October, with partial recovery by the end of the year. In fact, year-endwater levels were three feet lower than those observed in the beginning of the year. Thus, the

  • 11

    frequent groundwater level data from the transducers provide detailed information about thetiming of spring high and fall low water levels in the SMVMA. While it is locally understoodthat spring high levels in the SMVMA typically occur in late February to early March, thetransducer data indicate that the spring high level of 2015 in fact occurred in early January.

    The transducer data also indicate that the period of spring high groundwater levels in theSMVMA 2015, following the recovery of groundwater levels over winter 2014-2015, was verybrief. It is likely that seasonal agricultural irrigation commenced earlier than usual, perhaps dueto ongoing severe drought and/or changing irrigation practices, which would have contributed toan early decline in groundwater levels. Further, aquifer recharge from Sisquoc River discharge,which typically occurs over the winter months, was essentially nil.

    Given that a common objective of groundwater monitoring programs is to measure the springhigh and fall low groundwater levels, for example to facilitate estimation of annual change inbasin storage, the USGS came the closest of all agencies to meeting that objective in 2015 (withearly March and mid-October measurements). The transducer data provide the magnitude ofgroundwater level decline that occurred in 2015 in this portion of the SMVMA between earlyMarch and late April, from 5 to 10 feet. This in turn illustrates the magnitude of inconsistency inspring groundwater level data collected in the SMVMA by the USGS (early March), theSMVWCD (early April), and the SLOPWD (late April). Alternatively, the transducer dataindicate that groundwater levels declined to (and maintained) their fall low period during mid-September to mid-October 2015, thus illustrating the proper timing of fall measurements made inthe SMVMA (all agencies in that period).

    2.1.3 Groundwater Quality

    Groundwater quality conditions in the SMVMA have fluctuated greatly since the 1930's, whenhistorical water quality sampling began, with marked short- and long-term trends. Groundwaterquality in the SMVMA historically reflects the various natural sources of recharge to the aquifersystem, most notably streamflows of the Cuyama and Sisquoc Rivers that provide recharge alongthe Santa Maria River. The great majority of groundwater in the SMVMA, primarily in theeastern and central portions of the Santa Maria Valley and in the Sisquoc Valley, had historicallybeen of a calcium magnesium sulfate type originating from the Cuyama and Sisquoc Riverstreamflows. Further, groundwater was historically of better quality toward the Orcutt Upland,Nipomo Mesa, the City of Guadalupe, and coastal areas (Lippincott, J.B., 1931).

    With development of the Valley and surrounding areas in the 1940's through 1970's, includingexpansion of the agricultural and urban areas and addition of the Twitchell Reservoir project,groundwater quality conditions changed within the SMVMA. The changes includedimprovement of the general groundwater quality in the eastern to central part of the Santa MariaValley in and near the area of Twitchell Reservoir recharge, including the current-day municipalwellfield near the Town of Orcutt. Additionally, degradation of groundwater quality occurredfurther west and downgradient in the Valley, specifically with elevated general mineral andnitrate concentrations (USGS, Hughes, J.L., 1977).

  • 12

    Subsequently, from the 1970's through current day, general mineral concentrations ingroundwater have remained essentially unchanged, including the occurrence of better qualitywater in the SMVMA’s eastern, central, and southern portions and poorer quality water to thewest. Further, groundwater quality is generally slightly better and with less fluctuation in thedeep aquifer zone compared to the shallow, as shown by a map with representative historicalgroundwater quality graphs from areas throughout the SMVMA (Figure 2.1-5). Groundwaterquality data from 2015 indicate that total dissolved solids (TDS) concentrations in the shallowaquifer zone generally ranged between 700 and 1,100 mg/l in the Twitchell Recharge andMunicipal Wellfield Areas, and were about 650 to 1,000 mg/l in the Coastal Area. TDS valuesin the deep zone were slightly lower in the Twitchell Recharge Area (between 800 and 900 mg/l)and Municipal Wellfield Area (between 650 and 800 mg/l), but somewhat higher in the CoastalArea (between 700 and 1,400 mg/l). In portions of the deep coastal zone, TDS values also showa long-term gradual increase. TDS concentrations in the shallow aquifer zone of the SisquocValley are unknown but, in the deep zone, were about 600 mg/l. Overall, TDS values in theSMVMA generally remain at or below the California Department of Public Health’s secondarystandard of 1,000 mg/l.

    In contrast to the overall stability in general mineral (TDS) concentrations in groundwaterobserved during this recent period, nitrate concentrations in shallow groundwater haveprogressively increased. In 2015, nitrate-as-nitrate (nitrate-NO3) concentrations in shallowgroundwater remained elevated, in many areas above the primary drinking water standard of 45mg/l. In the Twitchell Recharge Area, nitrate concentrations were somewhat lower than thoseobserved in recent years, ranging from 5 to 20 mg/l (although a well typically with high nitrateconcentrations was not sampled in 2015). Nitrate concentrations in shallow groundwater in theMunicipal Wellfield Area continue a slight increasing trend from just above 50 mg/l a decadeago to about 65 mg/l currently. However, in the Coastal Area, nitrate concentrations in shallowgroundwater remained non-detect (less than 0.18 mg/l).

    Compared to widespread elevated nitrate concentrations in shallow groundwater, deepgroundwater concentrations remain markedly lower, generally less than 10 mg/l (nitrate-NO3).Extensive nitrate data from 2015 available from purveyor water supply wells (City of SantaMaria and GSWC), particularly for deep wells in the southern SMVMA, indicate nitrate-NO3values remain less than 10 mg/l. Exceptions to this observation have been two deeper wells inthe south-southeast part of the Valley (9N/33W-02A7 and 9N/34W-03F2), with nitrateconcentrations between 25 and 35 mg/l, and some coastal deep monitoring wells with nitratelevels exceeding 90 mg/l, as discussed below.

    Of particular importance to ongoing assessment of potential conditions of sea water intrusion arethe groundwater quality data from two sets of coastal monitoring wells. During an investigationconducted in the late 1960's, for which the monitoring well sets were constructed, localized areasof degraded shallow groundwater were identified but concluded at the time to be due toenvironmental factors other than intrusion (California DWR, 1970). Review of the coastalmonitoring results through 2015, in particular TDS values, provides an indication of whether seawater intrusion has occurred in the coastal SMVMA; review of coastal nitrate concentrationsprovides a measure of the extent and magnitude of water quality degradation from land use

  • 13

    activities further inland. Historical water quality graphs for these wells are provided inAppendix B.

    Since the commencement of coastal groundwater quality monitoring, coastal groundwater hascontinued to show elevated but largely unchanging specific conductance values. In 2015,shallow groundwater at the southerly monitoring well set (10N/36W-02Q, shallow well 02Q7,Figure 2.1-5) had a value of just under 1,000 mg/l; deep groundwater values (wells 02Q1, 02Q3,and 02Q4) have been lower, between 650 and 700 mg/l over the last 35 plus years. Groundwaterat the more northerly monitoring well set (11N/36W-35J) shows more variation in TDS valueswith depth: the wells with depths of 615, 495, and 228 feet have values of 700 mg/l, 1,300 mg/l,and 1,400 mg/l, respectively. TDS values in the shallowest well (35J5, 136 feet deep), havegradually risen throughout the monitoring period from about 1,000 mg/l in 1977 to 1,400 mg/l in2015.

    Some coastal groundwater, specifically in portions of the deep aquifer zone near the northerlymonitoring well set (11N/36W-35J), have shown gradually increasing degradation from nitrate,including through the present. Nitrate-NO3 concentrations have steadily risen from a range of 5to 10 mg/l in the 1980’s to between 40 and 95 mg/l in 2015 (see Figure 2.1-5). In contrast,groundwater in all aquifer zones near the southerly monitoring well set (10N/36W-02Q) haveconsistently shown very low concentrations of nitrate through the present. Shallow groundwatercontinued to have non-detectable levels of nitrate (less than 0.18 mg/l) and deep groundwaterconcentrations remained near 3 mg/l through 2015. Nitrate concentrations in the deepestgroundwater, specifically below a depth of 600 feet, along the coast (at both well sets) remainstable with values of around 3 mg/l or less.

    2.2 Twitchell Reservoir Operations

    In order to describe Twitchell Reservoir operations, monthly records of reservoir stage, storage,and releases were updated and recorded observations of reservoir conditions were noted. Thehistorical stage, storage, and releases, including through 2015, are described in relation toobserved climatic conditions in the SMVMA.

    2.2.1 Reservoir Stage and Storage

    Historical stage and storage in Twitchell Reservoir, for which reliable records begin in 1967,indicate a typical seasonal rise with winter and spring rain, followed by decline throughsubsequent spring and summer releases. Reservoir stage has risen to as high as about 640 feetmsl, corresponding to storage of nearly 190,000 acre-feet, on several occasions during the winterand spring months of years during which rainfall amounts were substantially higher thanaverage. Historical rises in stage have been rapid, occasionally over one or two months, withsubsequent declines gradually spread over the subsequent year or multiple years. During thoseyears when releases have essentially emptied the reservoir for purposeful supplementalgroundwater recharge through the Santa Maria River channel, the dam operator recorded theassociated minimum reservoir stage, which has risen over time from about 480 feet msl in 1968,to 525 feet MSL since 1986. This rise reflects the long-term filling of former dead pool storage(about 40,000 acre-feet below the reservoir outlet for release from conservation storage) with

  • 14

    sediment that has naturally occurred with operation of the project (SMVWCD, 1968-2015).These seasonal fluctuations and long-term rise in minimum stage, shown in relation to thereservoir conservation, flood control, and surcharge pools, are illustrated in a graph of historicalreservoir stage and storage (Figure 2.2.1a).

    It is noteworthy that the sedimentation of the former dead pool storage below the conservationoutlet in Twitchell Reservoir has not impeded the conservation of runoff for subsequent releasefor downstream groundwater recharge. Except for a few individual years over the life of thereservoir, accumulated storage in any year has been less than the designated active conservationpool of 109,000 af. In the infrequent wet years when greater storage could be conserved, e.g.1969, 1978, 1983, 1995, and 1998, the SMVWCD has been permitted to temporarily utilizesome of the dedicated flood control pool (89,000 af) to conserve those additional inflows andthen release them soon thereafter for downstream recharge. Total storage has never exceeded thecombined conservation pool and flood control pool storage volume (198,000 af) and has neverinvaded the uppermost surcharge pool (159,000 af above the conservation and flood controlpools) in the overall reservoir.

    Reservoir storage has historically risen to between 150,000 and nearly 190,000 acre-feet (af)during the winter and spring months of years during which rainfall was substantially higher thanaverage, with storage commonly below 50,000 af during most other years. As can be seen onFigure 2.2-1a, reservoir storage has repeatedly dropped to essentially zero during periods ofbelow-average rainfall, including those associated with drought conditions in 1976-77, 1987-90,and 2012-current year. Reservoir storage has also been essentially zero during most of 2000through 2004, 2007 through 2009, as a result of the overall drier climatic period that began in2002.

    Briefly during this drier climatic period, such as in both 2005 and 2006 when rainfall was aboveaverage, about 50,000 af of storage were accrued, all of which was released for downstreamgroundwater recharge. In late 2010 into early 2011, again in response to above-average rainfall,storage accrued by April 2011 to almost 93,000 af (and the stage to 615 feet MSL) with releasescommencing in February 2011 and continuing through March 2012. Since then, only a minoramount of water has been conserved that subsequently evaporated and/or was lost to seepagesuch that, during 2015, no releases were possible, reservoir stage remained at approximately 524feet MSL, and storage remained less than 1,000 af.

    2.2.2 Reservoir Releases

    Twitchell Reservoir annual releases for in-stream groundwater recharge since 1967 have rangedfrom zero during low rainfall years and drought periods to a maximum of 243,660 af in 1998, asillustrated in a bar chart of annual reservoir releases (Figure 2.2-1b). In general, and mostnotably in the Twitchell Recharge Area, groundwater levels have tended to track Twitchellreleases since the beginning of Reservoir operations (see Figure 2.1-2 and 2.2-1b). The long-term average annual release amount for the period 1967 through 2015 is 48,700 afy, with below-average releases during roughly two-thirds of those years. The five-year period from 1995through 1999 is notable for continual releases in amounts well above the annual average,reflecting a wetter climatic period from 1993 through 1998. Also notable are multiple year

  • 15

    periods when releases dropped to zero, specifically from 1987 through 1990 and from 2002through 2004, reflecting the drier climatic conditions during those periods of time. Whilereleases in 2005 and 2006 amounted to about 106,000 and 80,000 af, respectively, drier climaticconditions persisted with no releases for in-stream groundwater recharge in 2009 or 2010. Therelease of nearly 90,000 af of water from Twitchell Reservoir was conducted from Februarythrough December 2011, with the highest amounts during the months of June throughSeptember. In 2012, the beginning of the current severe drought, releases were well belowaverage, conducted only in January through March and amounting to only 9,100 af; since then,essentially no water has been available in storage and no releases have been made from thereservoir, including during 2015.Importantly, the magnitude of the effect on reservoir releases of the current dry climatic periodthat commenced in 2002 can be seen in a comparison of the calculated average reservoir releasesfor the following selected periods of time:

    Period of record 1967 – 2015 (48,700 afy);Pre-current dry period 1967 – 2001 (59,300 afy); andCurrent dry period 2002 – 2015 (22,100 afy).

    The average amount of water released during the current dry period (2002 – 2015) is less thanone-half the average for the overall period of record (1967 – 2015).

    2.3 Streams

    The surface water hydrology of the SMVMA is characterized in this section, specifically thecurrent conditions in relation to historical trends in stream discharge and quality.

    2.3.1 Discharge

    The main streams entering the SMVMA are the Cuyama and Sisquoc Rivers; these rivers join onthe Santa Maria Valley floor near Garey and become the Santa Maria River, which drains theValley from that point westward (see Figure 1.1-1 and Appendix A, Figure 3). The headwatersof the Sisquoc River include a portion of the San Rafael Mountains and Solomon Hills, and theRiver’s main tributaries within the SMVMA are Foxen, La Brea, and Tepusquet Creeks.Streamflow in the Sisquoc River and its tributary creeks have remained uncontrolled through thepresent. The Cuyama River drains a portion of the Sierra Madre Mountains, including theCuyama Valley, and streamflow into the Santa Maria River has been controlled sinceconstruction of Twitchell Dam between 1957 and 1959. The Santa Maria River receives minorstreamflows from two small tributaries, Suey and Nipomo Creeks, along its course toward theCity of Guadalupe and the Pacific Ocean. In the southern portion of the SMVMA, Orcutt Creekdrains a portion of the Solomon Hills (Solomon Canyon) and the Orcutt area, receivesintermittent flow from Graciosa Canyon, before ending near Betteravia.

    Stream discharge in the Cuyama River below the dam, recorded during the initial period ofTwitchell project operations between 1959 and 1983, averaged 37,350 afy. As discussed above,Twitchell Reservoir releases have averaged 48,700 afy from 1967 through 2015. The historicalvariation in reservoir releases and Cuyama River streamflow is shown in a bar chart of annual

  • 16

    surface water discharge for the River (Figure 2.3-1a). Cuyama River stream discharge, whichcomprises the largest source of SMVMA groundwater recharge, has ranged over the historicalperiod of record from no streamflow during several drought years, including as recently as 2010,2013, 2014, and 2015, to a high of almost 250,000 af during 1998.

    Stream discharge in the Sisquoc River, recorded at gauges at the southeast end of the Sisquocplain (“near Sisquoc”) and further downstream at the opening to the SMVMA (“near Garey”),averages 34,300 (absent data from years 1999-2007) and 36,400 afy, respectively, over thehistorical period of record.1 The downstream gauge (“near Garey”) provides a measure of thestream discharge entering the SMVMA from the Sisquoc plain, and it reflects inflow from theheadwaters of the Sisquoc River and its tributaries, as well as gains from and losses to theshallow aquifer in the Sisquoc plain. The historical variation in Sisquoc River streamflow isshown in a bar chart of annual surface water discharge for the River at both gauges (Figure 2.3-1b). Sisquoc River stream discharge to the SMVMA (“near Garey” gauge), which comprises alarge source of SMVMA groundwater recharge, has ranged over the historical period of recordfrom no streamflow during several drought years to over 300,000 af during 1998; the 2015annual discharge (provisional/approved) into the SMVMA was well below average, only about250 af. Of note is that the upstream gauge (“near Sisquoc”) was non-operational, and thus nodata are available, from 1999 through 2007. Further, discharge amounts in the tributaries Foxen,La Brea, and Tepusquet Creeks have not been recorded since the early 1970's (early 1980's forthe latter creek), when gauge operations were discontinued. As a result, the net amount ofgroundwater recharge in the Sisquoc plain from the Sisquoc River currently cannot be quantified.Reestablishment and monitoring of these currently inactive gauges (Foxen, La Brea, andTepusquet Creeks), as previously outlined in the SMVMA Monitoring Program andrecommended in this annual report, would provide for better understanding of the magnitude anddistribution of recharge from the Sisquoc River in the Sisquoc Valley.

    Streamflow in the Santa Maria River has been recorded at two gauges during varying periods oftime (see Appendix A, Figure 3). At the Guadalupe gauge, which was operational between 1941and 1987, stream discharge ranged from no streamflow during numerous years to almost 185,000af during 1941, and averaged 26,800 afy prior to the commencement of Twitchell projectoperations compared to 17,600 afy during the period of Twitchell project operations. Thehistorical variation in Santa Maria River streamflow is shown in a bar chart of annual surfacewater discharge for the River (Figure 2.3-1c). The reduction in streamflow at Guadalupe isattributed to Twitchell project operations, which are intended to maximize recharge along themore permeable portion of the River streambed by managing reservoir releases to maintain a“wetline” (downstream extent of streamflow) only as far as the Bonita School Road Crossing.

    Supplemental recharge to the Santa Maria Valley from Twitchell project operations has beenestimated to be about 32,000 afy based on comparison of pre- and post-project net losses instreamflow between Garey and Guadalupe (LSCE, 2000). The estimation does not account forchanges in climatic conditions between the pre-project (overall dry) and post-project (overallwet) periods or losses/gains along the Santa Maria River due to other processes, which couldresult in changes in the amount of water available for recharge over time. As a result ofdiscontinued stream discharge measurements at Guadalupe since 1987, combined with the lack

    1 These values of mean annual discharge include provisional (October through December) 2015 discharge data.

  • 17

    of gauged data for Suey and Nipomo Creeks, the net amount of groundwater recharge in theSanta Maria Valley from the Santa Maria River currently cannot be updated. Reestablishmentand monitoring of these currently inactive gauges (Suey Creek, Nipomo Creek, and Santa MariaRiver at Gaudalupe), as previously outlined in the SMVMA Monitoring Program andrecommended in this annual report, would provide for better understanding of the magnitude anddistribution of streamflow and recharge from the Santa Maria River.

    Stream discharge in the Santa Maria River has also been recorded more recently at a gauge atSuey Crossing northeast of the City of Santa Maria. However, these data are reported onlysporadically, as for years 1999 and 2006, or not at all, as in 2000 through 2005, and the dischargedata for 2009 through 2015 remain problematic. However, future acquisitions of the dischargedata from this gauge will also enhance an understanding of streamflow and recharge along theSanta Maria River.

    Stream discharge in Orcutt Creek, recorded at Black Road crossing from 1983 through thepresent (absent data from years 1992 through 1994), averages about 1,570 afy, ranging fromessentially no streamflow during several years to just over 10,000 af in 1995; in 2015, streamdischarge was well below average, approximately 220 af. The historical variation in streamflowis shown in a bar chart of annual surface water discharge for the creek (Figure 2.3-1d). Whileessentially all streamflow recorded at the gauge ultimately provides groundwater recharge to theSMVMA, it is not known how much groundwater recharge or discharge occurs upstream fromthe gauge, specifically between the gauge and the point where Orcutt Creek enters the SMVMA,or downstream from the gauge toward Guadalupe.

    2.3.2 Surface Water Quality

    The majority of recharge to the SMVMA has historically derived from streamflow in the SantaMaria River originating from the Cuyama and Sisquoc Rivers. Thus, groundwater quality inmuch of the SMVMA has historically reflected the water quality of streamflow in the Cuyamaand Sisquoc Rivers. Water quality in the rivers depends on the proportion and quality of therainfall runoff and groundwater inflow contributing to streamflow in their respective watershedsabove the Santa Maria Valley. The Cuyama River watershed includes the Cuyama Valley,which is reported to be underlain by geologic formations containing large amounts of gypsum;the Sisquoc River watershed is primarily steep terrain underlain by consolidated rocks (USGS,Worts, G.F., 1951).

    The quality of the streamflow in both the Cuyama and Sisquoc Rivers has historically been of acalcium magnesium sulfate type, although the Sisquoc River contains less sulfate and morebicarbonate than the Cuyama River. The Cuyama River quality has improved at two points intime during the historical period, specifically the mid-1940's and the late 1960's (USGS, Hughes,J.L., 1977). The improvement observed in the mid-1940's is thought to be due to agriculturaldevelopment of the Cuyama Valley that was supported by increased groundwater pumping inthat Valley for irrigation. The increased pumping lowered groundwater levels in the CuyamaValley, in turn reducing groundwater inflow to the Cuyama River, thereby reducing thecontribution of dissolved salts (sulfate in particular) to the River.

  • 18

    The improvement observed in the late 1960's is thought to be due to implementation of TwitchellReservoir project operations, which facilitated conservation of Cuyama River runoff andaugmented recharge to the Santa Maria Valley groundwater basin. Specifically, the higherstreamflow events in the Cuyama River that previously discharged to the ocean are of a betterquality due to dilution by greater rainfall runoff. Releases from Twitchell Dam therefore containlower concentrations of dissolved salts than the Cuyama River streamflows from the periodpreceding the project. The improvement in Cuyama River water quality from both of thesedevelopments may be seen in Table 2.3-1, which summarizes those earlier water quality resultsfrom the USGS (Hughes, J.L., 1977); more recent monitoring results from the USGS (1976 –2015) and the Central Coast Ambient Monitoring Program (CCAMP) (2000 – 2015), are alsoshown (CCRWQCB, 2015).

    Since operation of the Twitchell project began in the 1960s, Cuyama River water quality hasremained fairly constant. Reported TDS values range from about 650 to 1,200 mg/l; sulfate andchloride concentrations range from 190 to 760 mg/l, and from 25 to 85 mg/l, respectively.Nitrate-NO3 concentrations have remained low, ranging from

  • 19

    fluctuating between 500 and 3,000 mg/l, with values that exceeded 3,600 mg/l in 2005 and 2006.During the last decade, nitrate concentrations typically exceeded the health-based standard of 45mg/l, in fact exceeding 125 mg/l in 2007 through 2009 before declining to a range between 40and 70 mg/l since 2011. However, in 2015, the reported nitrate-NO3 concentration reached thehighest observed for this portion of Orcutt Creek, 135 mg/l.

    2.4 Climate

    The climatic data reported for the SMVMA are characterized in this section, specifically thecurrent conditions in relation to historical trends in precipitation and evapotranspiration data.

    2.4.1 Precipitation

    At least three precipitation gauges have historically been located in the SMVMA, at Guadalupe,Santa Maria (currently at the Airport and previously downtown), and Garey (see Appendix A,Figure 3). Additional gauges include two operated by the Santa Barbara County Public WorksDepartment at Sisquoc Ranch and Orcutt. The average annual rainfall measured at the SantaMaria Airport gauge, the most centrally located of the three gauges, is 12.63 inches, as shown ina bar chart of historical precipitation (Figure 2.4-1). Historically, the majority of rainfall occursduring the months of November through April. In calendar year 2015, the total rainfall was wellbelow average, at only 4.05 inches, with the greatest monthly amounts in December (1.12inches), November (0.69 inches), and April (0.67 inches), as shown in Table 2.4-1.

    Long-term rainfall characteristics for the SMVMA are reflected by the cumulative departurecurve of historical annual precipitation (on Figure 2.4-1), which indicates that the SMVMA hasgenerally experienced periods of wetter than normal conditions alternating with periods of drierthan normal to drought conditions. Wet conditions prevailed from the 1930's through 1944,followed by drier conditions from 1945 through the late 1960's. Subsequently, there have beenshorter periods of alternating wet and dry conditions, including the most recent cycle of a wetperiod in the early-1990's to 1998, followed by the current overall dry period from 2002 through2015. Conditions have shown short-term variation with rainfall totals above the long-termaverage in 2010 and 2011 but well below the average since 2012. This pattern of fluctuations inclimatic conditions closely corresponds to the long-term fluctuations in groundwater levelsdescribed in Section 2.1.2 above, including the substantial decline observed between 1945 andthe late 1960's and the subsequent repeating cycle of decline and recovery between historical-lowand historical-high groundwater levels. Most recently, groundwater levels rose substantially inmuch of the SMVMA through 2011 in response to large amounts of rainfall in late 2010 andearly 2011 (and the associated recharge from prolonged Twitchell Reservoir releases and highSisquoc River discharge). However, the overall decline in groundwater levels observed since2002 and particularly since 2012 is primarily attributed to the continued below average rainfall,Twitchell releases, and Sisquoc River discharge.

    Importantly, the magnitude of the effect on rainfall amounts of the current dry climatic periodthat commenced in 2002 can be seen in a comparison of the calculated average precipitation forthe following selected periods of time. They include the entire period of record (1948 – 2015) as

  • 20

    well as the same periods as those for historical Twitchell Reservoir releases described in Section2.2.2 above):

    Precipitation data period of record 1948 – 2015 (12.63 inches)Twitchell period of record 1967 – 2015 (13.14 inches);Twitchell pre-current dry period 1967 – 2001 (13.95 inches); andCurrent dry period 2002 – 2015 (11.10 inches).The average rainfall during the current dry period (11.10 inches for 2002 – 2015) is 84 percent ofthe average for the Twitchell period of record (13.14 inches for 1967 – 2015) and 88 percent ofthe average for the precipitation data period of record (12.63 inches for 1948 – 2015).

    2.4.2 Evapotranspiration

    Three CIMIS climate stations were initially operated within the SMVMA for varying periods oftime, specifically at Santa Maria, Betteravia, and Guadalupe between 1983 and 1997 (seeAppendix A, Figure 3). Subsequently, CIMIS stations began operating near Sisquoc and on thesouthern Nipomo Mesa, the latter located just outside of the SMVMA, with climate dataavailable for full calendar years beginning in 2001 and 2007, respectively. Most recently, aCIMIS climate station located on the floor of the Santa Maria Valley (“Santa Maria II” near theSanta Maria airport, see Appendix A, Figure 3) was reestablished in April 2011. A full calendaryear of data from Santa Maria II was available for the first time in 2012. These six stations haverecorded daily reference evapotranspiration (ETo) and precipitation amounts, with annual ETovalues typically ranging between 42 and 53 inches and averaging about 48 inches, as shown in abar chart of the historical ETo values for the SMVMA (Figure 2.4-2).

    Daily climate data for 2015 from the Santa Maria II, Nipomo, and Sisquoc stations are listed inTable 2.4-2, specifically daily, monthly, and annual ETo and precipitation amounts. Annual ETovalues ranged from 43.49 inches (Nipomo) to 51.49 inches (Sisquoc). Annual precipitationamounts were reported as 5.61 inches at the Sisquoc station but had questionable recorded totalsof 1.65 inches at the Santa Maria II station and 8.29 inches at the Nipomo station.

    Several characteristics of the 2015 CIMIS station data are worthy of note. Evapotranspirationwas highest during the months of March through September at all three stations, and the ETovalues from the Santa Maria II station were typically intermediate to those from the other twostations. In addition, the 2015 total precipitation recorded at the Sisquoc CIMIS station was themost similar to the total amount observed at the Santa Maria Airport precipitation gauge.However, the monthly precipitation amounts reported at all three CIMIS stations do not trackthose of the airport gauge. Similar inconsistencies in precipitation data between CIMIS stationsand the airport gauge have been observed since 2012. For this reason, and as described in thenext chapter, the 2015 ETo data from the Santa Maria II CIMIS station and the 2015precipitation data from the airport gauge were utilized in the estimation of agricultural waterrequirements for the SMVMA in 2015.

  • Qa

    l

    Qa

    l

    Qa

    l

    ?

    B'

    B

    A

    A'

    Qa

    l

    Qd

    Tp

    r

    Qal

    Qa

    l

    Qs

    09

    N/3

    4W

    10

    N/3

    4W

    09

    N/3

    5W

    09

    N/3

    3W

    10

    N/3

    3W

    10

    N/3

    5W

    09

    N/3

    2W

    10

    N/3

    2W

    11

    N/3

    2W

    11

    N/3

    3W

    11N

    /34

    W11

    N/3

    5W

    09

    N/3

    1W

    10

    N/3

    1W

    11

    N/3

    1W

    10

    N/3

    6W

    08

    N/3

    5W

    08

    N/3

    4W

    08

    N/3

    3W

    08

    N/3

    2W

    08

    N/3

    1W

    11

    N/3

    6W

    09

    N/3

    6W

    08

    N/3

    6W

    Fig

    ure

    2.1-

    1aG

    ener

    aliz

    edG

    eolo

    gic

    Map

    wit

    hC

    ross

    Sec

    tion

    Loc

    atio

    ns

    San

    taM

    aria

    Val

    ley

    Man

    agem

    ent

    Are

    a

    02

    41

    Mile

    s

    Leg

    en

    d

    !

    !Cro

    ss

    Se

    ctio

    nL

    oca

    tio

    n

    Ma

    na

    ge

    men

    tA

    rea

    Bo

    un

    da

    ry

    Gro

    und

    wa

    ter

    Ba

    sin

    Bo

    un

    da

    ry

    Su

    rfic

    ial

    Ge

    olo

    gy

    Dun

    eS

    an

    d

    Qu

    ate

    rna

    ryA

    lluvi

    um

    Old

    er

    Du

    ne

    Sa

    nd

    Te

    rra

    ceD

    ep

    osits

    Orc

    utt

    Fo

    rma

    tion

    Pa

    soR

    ob

    les

    Fm

    an

    dC

    are

    ag

    aS

    and

    Te

    rtia

    ryU

    nd

    iffe

    ren

    tia

    ted

    !

    Qs

    Qal

    Qd

    Qt

    Qo

    Tpr

    Tu

    SA

    NT

    AM

    AR

    IA

    VA

    LL

    EY

    SA

    NT

    AM

    AR

    IA

    VA

    LL

    EY

    NIP

    OM

    ON

    IP

    OM

    OM

    ES

    AM

    ES

    A

    NO

    RT

    HE

    RN

    NO

    RT

    HE

    RN

    CIT

    IE

    SC

    IT

    IE

    S

  • !

    !

    "!

    "

    !

    "

    !

    "

    "!

    "

    !

    09N

    /34W

    10N

    /34W

    09N

    /35W

    09N

    /33

    W

    10N

    /33W

    10N

    /35W

    09N

    /32W

    10N

    /32W

    11N

    /32

    W11

    N/3

    3W

    11N

    /34

    W11

    N/3

    5W

    09N

    /31

    W

    10N

    /31W

    11N

    /31

    W

    10N

    /36W

    08N

    /35W

    08N

    /34

    W08N

    /33

    W08N

    /32W

    08N

    /31

    W

    11N

    /36

    W

    09N

    /36W

    08N

    /36W

    33

    G1

    28M

    1

    02Q

    702Q

    311

    E4

    06

    N1

    /06N

    3

    30G

    1

    19B

    1

    09R

    108

    H1

    12

    R2

    23

    K1

    Fig

    ure

    2.1

    -2H

    isto

    rica

    lG

    rou

    nd

    wat

    erL

    evel

    sS

    anta

    Mar

    iaV

    all

    eyM

    anag

    emen

    tA

    rea

    05

    10 M

    iles

    "

    Ma

    pL

    eg

    en

    d

    !S

    ha

    llo

    wW

    ell

    "D

    ee

    pW

    ell

    SM

    VM

    A

    Gra

    ph

    Le

    ge

    nd

    "S

    ha

    llow

    Gro

    un

    dw

    ate

    rL

    eve

    l

    "D

    eep

    Gro

    un

    dw

    ate

    rL

    eve

    l

    Gro

    un

    dS

    urf

    ace

    0

    Sis

    qu

    oc

    Va

    lle

    yS

    is

    qu

    oc

    Va

    lle

    y0

    0

    0

    Mu

    nic

    ip

    al

    We

    llf

    ie

    ld

    Ar

    ea

    Mu

    nic

    ip

    al

    We

    llf

    ie

    ld

    Ar

    ea

    Flo

    win

    g-

    No

    tM

    ea

    su

    red

    00

    Co

    as

    ta

    lA

    re

    aC

    oa

    st

    al

    Ar

    ea

    00

    Os

    oF

    la

    co

    Ar

    ea

    Os

    oF

    la

    co

    Ar

    ea

    " %

    0

    0

    Ce

    nt

    ra

    lA

    gr

    ic

    ult

    ur

    al

    Ar

    ea

    Ce

    nt

    ra

    lA

    gr

    ic

    ult

    ur

    al

    Ar

    ea

    Tw

    it

    ch

    ell

    Re

    ch

    ar

    ge

    Ar

    ea

    Tw

    it

    ch

    ell

    Re

    ch

    ar

    ge

    Ar

    ea 0

    Gro

    un

    dS

    urf

    ace

    Ele

    va

    tion

    =3

    23'

  • #

    ##

    #

    #

    #

    #

    #

    #

    #

    #

    #

    #

    #

    #

    !$

    !$

    !$

    !$

    !$

    !$

    !$

    !$

    !$

    !$

    !$

    !$

    !$

    !$

    !$

    !$

    !$

    !$

    !$

    !$

    !$

    !$

    !$

    !$

    !$

    !$

    !$

    !$

    !$

    !$

    09

    N/3

    4W

    10

    N/3

    4W

    09

    N/3

    5W

    09

    N/3

    3W

    10

    N/3

    3W

    10

    N/3

    5W

    09

    N/3

    2W

    10

    N/3

    2W

    11N

    /32

    W11

    N/3

    3W

    11N

    /34

    W11

    N/3

    5W

    09

    N/3

    1W

    10

    N/3

    1W

    11N

    /31

    W

    10

    N/3

    6W

    08

    N/3

    5W

    08

    N/3

    4W

    08

    N/3

    3W

    08

    N/3

    2W

    08

    N/3

    1W

    11N

    /36

    W

    09

    N/3

    6W

    22

    C2

    5

    09

    F1

    35

    12

    C1

    *7

    33

    G3

    36

    23

    G1

    32

    33

    J17

    2

    24

    Q1

    56

    20

    H3

    60

    14

    E4

    63

    09

    D1

    56

    06

    N3

    55

    08

    H1

    99

    02

    Q7

    *1

    6

    13

    J11

    21

    35

    B1

    25

    2

    28

    A1

    19

    8

    27

    G1

    20

    6(S

    )

    21

    P1

    17

    8

    20

    H1

    15

    9

    18

    H1

    14

    8

    16

    L1

    18

    8

    07

    R6

    13

    8

    07

    M1

    111

    24

    L1

    32

    9

    22

    L_

    29

    2

    23

    K1

    50

    22

    2D

    14

    20

    16

    L1

    36

    8

    06

    D1

    27

    7

    07

    R1

    14

    1

    Hki

    wtg

    !3/2

    .4c

    Eqpvq

    wtu

    !qh!

    Gs

    wcn

    !Itq

    wpfy

    cvgt

    !Gng

    xcvk

    qp-!Ujcn

    nqy

    !\qp

    g-!G

    ctn{

    !Urtk

    pi!

    )Oct

    ej!;

    !.!O

    ctej

    !38*!

    3126

    Ucp

    vc!O

    ctkc

    !Xcnng{

    !Ocp

    cigo

    gpv!

    Ctg

    c

    02

    41

    Mile

    s

    Ma

    pL

    eg

    en

    d

    !$

    Sha

    llow

    We

    ll

    #U

    ncla

    ssi

    fie

    dW

    ell

    Gro

    und

    wa

    ter

    Ele

    vatio

    nC

    on

    tou

    r(f

    t,N

    AV

    D8

    8)

    Sa

    nta

    Ma

    ria

    Va

    lley

    Ma

    nag

    em

    entA

    rea

    Bo

    un

    da

    ry

    "

    Gro

    un

    dw

    ate

    rE

    leva

    tio

    nD

    ata

    used

    for

    Co

    nto

    uri

    ng

    We

    llID

    Da

    teR

    PE

    DT

    WW

    SE

    Ag

    en

    cy

    09N

    /32W

    -06D

    13

    /10/2

    015

    437

    .716

    1.10

    27

    7U

    SG

    S

    09N

    /32W

    -16L

    13

    /10/2

    015

    470

    .610

    2.3

    336

    8U

    SG

    S

    09N

    /32W

    -22D

    13

    /10/2

    015

    497

    .777.4

    142

    0U

    SG

    S

    09N

    /32W

    -23K

    13

    /10/2

    015

    534

    .732

    .48

    50

    2U

    SG

    S

    09N

    /33W

    -22L

    _3/2

    7/2

    015

    85

    255

    9.70

    29

    2T

    MA

    09N

    /33W

    -24L1

    3/9

    /2015

    533

    .720

    4.66

    32

    9U

    SG

    S

    09N

    /34W

    -08H

    13/1

    1/2015

    224

    .712

    5.6

    39

    9U

    SG

    S

    10N

    /33W

    -07

    M1

    3/1

    6/2

    015

    257

    .714

    6.6

    811

    1U

    SG

    S

    10N

    /33

    W-0

    7R

    13

    /13/2

    015

    272

    .713

    1.80

    141

    US

    GS

    10N

    /33W

    -07

    R6

    3/1

    7/2

    015

    272

    .713

    4.9

    813

    8U

    SG

    S

    10N

    /33W

    -16

    L1

    3/1

    2/2

    015

    297.4

    910

    9.4

    318

    8U

    SG

    S

    10N

    /33W

    -18

    H1

    3/1

    7/2

    015

    278

    .713

    0.4

    414

    8U

    SG

    S

    10N

    /33

    W-2

    0H

    13/1

    2/2

    015

    302

    .714

    3.8

    515

    9U

    SG

    S

    10N

    /33W

    -21P

    13/1

    2/2

    015

    316

    .713

    8.6

    317

    8U

    SG

    S

    10N

    /33

    W-2

    7G

    13/1

    1/2

    015

    340

    .713

    4.3

    0206

    (S)

    US

    GS

    10N

    /33

    W-2

    8A

    13/1

    1/2

    015

    327

    .713

    0.1

    019

    8U

    SG

    S

    10N

    /33

    W-3

    5B

    13/1

    1/2

    015

    352

    .710

    0.6

    125

    2U

    SG

    S

    10N

    /34

    W-0

    6N

    33/9

    /2015

    156

    .710

    1.2

    95

    5U

    SG

    S

    10N

    /34

    W-0

    9D

    13/9

    /2015

    185

    .712

    9.6

    55

    6U

    SG

    S

    10N

    /34W

    -13

    J13/1

    3/2

    015

    262

    .714

    1.8

    112

    1U

    SG

    S

    10N

    /34

    W-1

    4E

    43/1

    6/2

    015

    222

    .715

    9.2

    86

    3U

    SG

    S

    10N

    /34

    W-2

    0H

    33/9

    /2015

    182

    .712

    2.8

    06

    0U

    SG

    S

    10N

    /35

    W-0

    9F

    13/9

    /2015

    90.7

    55

    .75

    35

    US

    GS

    10N

    /35

    W-2

    4Q

    13/1

    1/2

    015

    164

    .710

    9.0

    75

    6U

    SG

    S

    10N

    /36

    W-0

    2Q

    7*

    11/1

    9/2

    014

    17.9

    1.75

    16U

    SG

    S

    11N

    /34

    W-3

    3J1

    3/9

    /2015

    192

    .712

    0.3

    372

    US

    GS

    11N

    /35

    W-2

    2C

    23/1

    6/2

    015

    241.

    5236.9

    75

    Wo

    odla

    nds

    11N

    /35

    W-2

    3G

    13/1

    2/2

    015

    257.3

    22

    5.1

    032

    TM

    A

    11N

    /35

    W-3

    3G

    33/1

    2/2

    015

    94.7

    58

    .50

    36

    TM

    A

    11N

    /36

    W-1

    2C

    1*4/9

    /201

    521.

    414

    .177

    Phill

    ips

    66

  • !$

    !$

    !$

    !$

    !$

    !$

    !$

    !$ !$

    !$

    !$

    !$

    !$

    !$

    !$

    !$

    !$

    !$

    !$

    !$!$

    !$

    !$

    !$

    09

    N/3

    4W

    10

    N/3

    4W

    09

    N/3

    5W

    09

    N/3

    3W

    10

    N/3

    3W

    10

    N/3

    5W

    09

    N/3

    2W

    10

    N/3

    2W

    11N

    /32

    W11

    N/3

    3W

    11N

    /34